Environ Sci Pollut Res DOI 10.1007/s11356-009-0258-6

RESEARCH ARTICLE

Atmospheric heavy metal deposition in Northern Vietnam: Hanoi and Thainguyen case study using the moss biomonitoring technique, INAA and AAS Hung Nguyen Viet & Marina Vladimirovna Frontasyeva & Thu My Trinh Thi & Daniel Gilbert & Nadine Bernard

Received: 19 May 2009 / Accepted: 27 October 2009 # Springer-Verlag 2009

Abstract Background, aim, and scope The moss technique is widely used to monitor atmospheric deposition of heavy metals in many countries in Europe, whereas this technique is scarcely used in Asia. To implement this international reliable and cheap methodology in the Asian countries, it is necessary to find proper moss types typical for the Asian environment and suitable for the biomonitoring purposes. Such a case study was undertaken in Vietnam for assessing Responsible editor: Henner Hollert M. V. Frontasyeva : T. M. Trinh Thi (*) Department of Neutron Activation Analysis, Frank Laboratory of Neutron Physics, Joint Institute for Nuclear Research, 141980 Dubna, Moscow Region, Russia e-mail: [email protected] M. V. Frontasyeva e-mail: [email protected] H. Nguyen Viet Department of Public Health and Epidemiology, Swiss Tropical Institute, Socinstrasse 57, P.O. Box 4002, Basel, Switzerland e-mail: [email protected] N. Bernard Laboratoire de Chrono-environnement, UMR UFC/CNRS 6249 USC INRA, Université de Franche-Comté, Place Leclerc, 25 000 Besancon cedex, France D. Gilbert Laboratoire de Chrono-environnement, UMR UFC/CNRS 6249 USC INRA, Université de Franche-Comté, Pôle Universitaire du Pays de Montbéliard, 4 place Tharradin, BP 71427, 25 211 Montbeliard cedex, France e-mail: [email protected]

the environmental situation in strongly contaminated areas using local species of moss Barbula indica. Materials and methods The study is focused on two areas characterized by different pollution sources: the Hanoi urban area and the Thainguyen metallurgical zone. Fifty-four moss samples were collected there according to standard sampling procedure adopted in Europe. Two complementary analytical techniques, atomic absorption spectrometry (AAS) and instrumental neutron activation analysis (INAA), were used for determination of elemental concentrations in moss samples. To characterize the pollution sources, multivariate statistical analysis was applied. Results and discussion A total of 38 metal elements were determined in the moss by the two analytical techniques. The results of descriptive statistics of metal concentration in moss from the city center and periphery of Hanoi determined by AAS are presented. The similar results for moss from Thainguyen province determined by INAA and AAS are given also. A comparison of mean elemental concentrations in moss of this work with those in different environmental conditions of other authors provides reasonable information on heavy metal atmospheric deposition levels. Factor loadings and factor scores were used to identify and apportion contamination sources at the sampling sites. The values of percentage of total of factors show two highly different types of pollution in the two examined areas—the Hanoi pollution composition with high portion of urban-traffic activity and soil dust (62%), and the one of Thainguyen with factors related to industrial activities (75%). Besides, the scatter of factors in factor planes represents the greater diversity of activities in Hanoi than in Thainguyen. Conclusions Good relationship between the result of factor analysis and the pollution sources evidences that the moss

Environ Sci Pollut Res

technique is a potential method to assess the air quality in Vietnam. Recommendations and perspectives Moss B. indica widely distributed in Vietnam and Indo-China is shown to be a reliable bryophyte for biomonitoring purposes in sub-tropic and tropic climate. However, the necessity of moss interspecies calibration is obvious for further studies in the area to provide results compatible with those for other Asian countries and Europe.

Thainguyen (Tran 2001). The local epiphytic moss Barbula indica widely spread in Vietnam and abundant throughout tropical Asia (Eddy 1990) as well as in other tropical and sub-tropical regions (Daly et al. 2006) has been used in this study.

2 Materials 2.1 Study areas

Keywords Moss . Metal deposition . Biomonitoring . Barbula indica . AAS . INAA . Vietnam

1 Background, aim and scope In Vietnam, the environmental quality has been gradually degraded by the new economic policies, which promote increasing industrialization, urbanization, construction, traffic, as well as agricultural development (Pham et al. 1995). For a long time, environmental quality monitoring has often been neglected in the country. While several studies assessing metal pollution in aquatic environments have been carried out (Buschmann et al. 2008; TarrasWahlberg et al. 2008; Marcussen et al. 2008), metal atmospheric environment monitoring data have remained limited and often focused on large cities such as Hanoi and Ho Chi Minh city (Pham et al. 1999, 2004). Bryophytes (moss and liverwort) lack a cuticle and a root system and obtain nutrients as particulates and in solution directly from atmospheric deposition. They have good bioaccumulating ability, particularly for heavy metals, where metal concentrations reflect deposition without the complication of additional uptake via a root system (Mulgrew and Williams 2000). Moss is often used to monitor atmospheric metal pollution as well as other trace elements in Europe, whereas this technique is scarcely used in Asia (China: Xiao et al. 1997; Lee et al. 2005; Bangladesh: Tarafdar et al. 1997) and has never been used in Vietnam. The moss technique first introduced in Scandinavia has proved to be feasible for studying atmospheric deposition of heavy metals as well as other trace elements (Ruhling 2002). Heavy metals accumulate in moss due to their morphological and physiological characteristics. Moss leaves are highly permeable for ions of trace elements and mineral salts. Ions found in moss come from atmospheric precipitation and dry deposition through simple processes of ion exchange, while the uptake from substratum is negligible (Steinnes 1993). The recommended species of mosses for metal deposition monitoring in Europe are Pleurozium schreberi, Hylocomium splendens, and Scleropodium purum (Reynolds et al. 2001). However, these species are not found in Hanoi and

Hanoi (21.02°N, 105.85°E) is the second largest city of Vietnam (after Ho Chi Minh City) with a population of about 3.5 million habitants. Recently, Hanoi has become larger by including the province of Hatay. The study surface of 912 km2 was divided into two zones (Fig. 1), corresponding to urban zone 1 (82 km2, i.e., 9%) and rural zone 2 (830 km2, i.e., 91%). Indeed, the urban area of Hanoi, composed of seven inner city districts at the moment of this study, shows a high population density (i.e., 53% of population), heavy traffic, and some mediumsized industrial zones. The rural area is composed of four surrounding districts where agriculture is the principal activity with some small pre-existing industrial zones and a few of larger ones that are currently being built. In these surrounding districts, agricultural activities can be considered as the main pollution source via pesticides and chemical fertilization (Nguyen 1996; ADB 2002). Thainguyen province (21.55°N, 105.87°E) located 80 km north of Hanoi (Fig. 1) has an area of 3,541 km2 and a population of more than one million. Thainguyen for a long time has been a center of metallurgical industry with a closed metallurgical production line from iron ore to steel making, casting, and rolling (Thainguyen Iron and Steel Corporation has been established since 1959). In Thainguyen, the urban life is connected with industrial activities, and a part of them takes place inside the city. Ironsteel industry and other mineral exploitation (coal and nonferrous metals as lead, zinc, tungsten, gold, copper, mercury, and tin) are the main industrial activities of Thainguyen and the main pollution sources of the area, while the rest of the province remains rural engaged in agriculture and forestry. The climate in Hanoi during summer is characterized by elevated temperature and high precipitation. During the sampling period from April to June of 2002, the minimal temperature varied from 16.8°C to 25.0°C, the maximal from 33.6°C to 38.6°C, and the average temperature was 27.6±2.1°C. The humidity varied from 80.0% to 82.0% (mean, 81.0±1.0%). The rainfall during this period was from 59.0 to 240.0 mm (mean, 171.0±97.9 mm). The total evaporation in this period was from 79.0 to 86.0 mm (mean, 83.3±3.8 mm; VNJMS 2002). The climate in Thainguyen during sampling period in October of 2003

Environ Sci Pollut Res

b

Thainguyen

Northern Vietnam

a

Hanoi

Fig. 1 Sampling areas and sites. a Hanoi sampling sites (filled circle, zone 1 sites; open circle, zone 2 sites). b Thainguyen sampling sites

was rather similar to that of Hanoi with the average temperature of 25.3±1.7°C, the mean humidity of 76.8± 5.8%, and the average rainfall of 66±52 mm. 2.2 Sampling A total of 29 moss samples were taken at different sites of the city of Hanoi, comprising 19 sites in zone 1 and ten sites in zone 2 (Fig. 1). A sampling grid of 2×2 km was used in zone 1 and of 10×10 km for zone 2. Moss samples were collected from April to June of 2002. In Thainguyen province, 25 sampling sites were selected around the smelter zone at different distances from it (Fig. 1). The moss sampling was carried out in October 2003. Moss sampling was carried out according to the protocol defined in the guidelines of the program “European survey of atmospheric heavy metal deposition” (Reynolds et al. 2001). However, in some cases of our study, sampling sites were not far enough from roads, in particular, in the city center. As B. indica is an acrocarpous species, with strong

contact to the substrate, the following sampling way was carried out to minimize the influence of substrate. At each site, five to ten sub-samples collected on hard substrate and in non-trampled places (walls and roofs, tree trunks) were combined to one collective sample. The moss B. indica is a small bryophyte, green or yellowish, forming low tufts with evenly distributed stems seldom exceeding 1 cm in height. In the laboratory, moss samples were cleaned from extraneous materials (large soil particles, leaves, dust, etc.). The top part of the mosses (living, green) was separated from the lower part (brown, dead), and then only the top part was used for analysis.

3 Methods 3.1 Atomic absorption spectrometry (AAS) Determination of seven elements (Zn, Mn, Fe, Cd, Ni, Cu, and Pb) using atomic absorption spectrometry (AAS) was

Environ Sci Pollut Res Table 1 Flux parameters of irradiation channels Irradiation channel Φ·1012, n⋅cm−2⋅s−1 E=0–0.55eV Φ·1012, n⋅cm−2⋅s−1 E=0.55–105 eV Φ·1012, n⋅cm−2⋅s−1 E=105–25⋅106 eV Temperature Thermal Resonance Fast (°C) Ch1 Ch2

(Cd-screened) 0.02 1.2

3.3 3.0

carried out at the Laboratory of Environmental Biology, University Franche-Comté, Besançon, France. For AAS, the unwashed moss samples were dried at 40°C to constant weight before the analysis. About 150 mg of dry mass was ground, digested in a mixture of 3.4 ml of concentrated nitric acid (HNO3, 65%), and 0.6 ml of H2O2 at 65°C during 48 h. The resulting solution was diluted in 16 ml of deionized water and filtered through 0.2µm mesh. The heavy metal concentrations in the moss were determined using furnace (for Pb, Cd, Ni, and Cu) and flame (for Zn, Mn, and Fe) atomic absorption spectrometer VARIAN® 220Z. The quality assurance of AAS was checked by measuring concentrations of these seven elements in the reference material ray grass (CRM 281, No. 766 of the Community Bureau of Reference, Commission of the European Community). Analytical replication was used for each analysis; the same reference material was used after every tenth analysis and during the whole period. 3.2 Instrumental neutron activation analysis (INAA) The instrumental neutron activation analysis (INAA) was carried out at the Department of Neutron Activation Analysis, Frank Laboratory of Neutron Physics, Joint Institute for Nuclear Research, Dubna, Russia. For INAA, about 0.3 g dry weight of moss sample was heat-sealed in polyethylene foil bags for short-term irradiation and packed in aluminum cups for long-term irradiation. Samples were

4.2 4.1

70 60

irradiated in channels equipped with a pneumatic system in the pulsed fast reactor IBR-2, FLNP, JINR. Neutron flux density characteristics and temperature are given in Table 1 (Frontasyeva et al. 2000). To determine short-lived isotopes, every sample was irradiated for 3–5 min at channel 2 (Ch2) then measured twice for 5–8 and 20 min after 3–5 min of decay. To determine elements yielding long-lived isotopes, reference materials packed together with ten to 12 samples in a container were irradiated for 5 days at the Cd-screened channel 1 (Ch1). Then, the reference materials and samples were re-packed and measured twice after 4–5 and 20 days of decay; the duration of measurements was 45 min and 2 h 30 min, respectively. Sample gamma spectra were obtained by measuring with a Ge(Li) detector with a resolution of 2.5–3 KeV for the 60Co 1,332 KeV line or a HPGe detector with a resolution of 1.9 KeV for the 60Co 1,332 KeV line. Data processing was performed using the software developed in FLNP JINR, and element contents were determined on the basis of certified reference materials and flux comparators (Frontasyeva et al. 2000). To provide quality control, contents of elements yielding short- and long-lived isotopes were determined using certified reference materials: (RM) Lichen 336 IAEA (International Atomic Energy Agency) and standard reference materials SRM-1575 (Pine Needles) from the US NIST (National Institute of Standards and Technology).

Table 2 Metal concentrations (µg.g−1) in moss Barbula indica from the city center and periphery of Hanoi determined by atomic absorption spectrometry Zone 1 (city center, n=19)

Pb Cd Zn Ni Cu Mn Fe a

p valuea

Zone 2 (periphery, n=10)

Min

Max

Median

Mean

SE

Min

Max

Median

Mean

SE

10 0.4 86 9.4 16 92 1,870

56 2.7 1,000 38 80 391 12,900

31 1.0 300 16 30 198 6,920

32 1.2 360 19 37 207 6,820

13 0.7 230 7.8 17 76 2,870

6 0.3 70 8.6 12 97 2,280

40 1.5 1,290 21 70 404 13,900

11 0.7 110 12 24 214 5,710

14 0.7 230 13 27 246 6,860

10 0.3 370 4 16 101 3,480

Mann–Whitney test

SE standard error, NS non-significant, n number of samples